polish-blog
Pass
Audited by Gen Agent Trust Hub on Jun 19, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill is susceptible to Indirect Prompt Injection (Category 8) as it processes untrusted user-provided blog content.
- Ingestion points: The skill reads the entire content of a blog article provided in the agent's context (referenced in Phase 1 and Phase 2).
- Boundary markers: No specific delimiters or instructions to ignore embedded commands within the blog content are defined.
- Capability inventory: The skill utilizes web search tools to find references (Phase 3) and performs file system write operations to update the blog post (Phase 5).
- Sanitization: No explicit sanitization or validation of the input article content is mentioned.
- Note: The risk is significantly mitigated by the 'Confirm and Apply' phase (Phase 5), which mandates that the agent present all suggestions to the user and wait for explicit approval before executing any file writes.
Audit Metadata